Thermodynamics of Ni2+, Cu2+, and Zn2+ Binding to the Urease

Aug 21, 2007 - Organisms, therefore, invest genetic and biochemical re- sources to ensure that ... urease have met with little success (5), and active...
1 downloads 0 Views 304KB Size
10506

Biochemistry 2007, 46, 10506-10516

Thermodynamics of Ni2+, Cu2+, and Zn2+ Binding to the Urease Metallochaperone UreE† Nicholas E. Grossoehme,‡ Scott B. Mulrooney,§ Robert P. Hausinger,§,| and Dean E. Wilcox*,‡ Department of Chemistry, Dartmouth College, HanoVer, New Hampshire 03755, and Department of Microbiology and Molecular Genetics and Department of Biochemistry and Molecular Biology, Michigan State UniVersity, East Lansing, Michigan 48824 ReceiVed January 26, 2007; ReVised Manuscript ReceiVed June 22, 2007

ABSTRACT: The two Ni2+ ions in the urease active site are delivered by the metallochaperone UreE, whose metal binding properties are central to the assembly of this metallocenter. Isothermal titration calorimetry (ITC) has been used to quantify the stoichiometry, affinity, and thermodynamics of Ni2+, Cu2+, and Zn2+ binding to the well-studied C-terminal truncated H144*UreE from Klebsiella aerogenes, Ni2+ binding to the wild-type K. aerogenes UreE protein, and Ni2+ and Zn2+ binding to the wild-type UreE protein from Bacillus pasteurii. The stoichiometries and affinities obtained by ITC are in good agreement with previous equilibrium dialysis results, after differences in pH and buffer competition are considered, but the concentration of H144*UreE was found to have a significant effect on metal binding stoichiometry. While two metal ions bind to the H144*UreE dimer at concentrations 99% pure and obtained from Sigma. KaUreE and H144*UreE samples were purified from Escherichia coli C41(DE3) cells containing pETWT or pETH144*, respectively, as described previously (14, 15). The pACT-BpEFGD construct used for BpUreE expression in Escherichia coli C41(DE3) cells is described elsewhere (23); the protein was purified by standard chromatographic methods using Ni-nitrilotriacetic acid, Superdex-75, and Q-Sepharose resins (GE Healthcare). Protein concentrations were determined spectrophotometrically, based on molar extinction coefficients at 280 nm calculated from the protein sequences (24): 5120 M-1 cm-1 for H144*UreE and KaUreE, and 21 430 M-1 cm-1 for BpUreE. ITC experiments were carried out at 25 ( 0.2 °C, unless indicated otherwise, on a MicroCal VP-ITC calorimeter, as described previously (25). Samples were buffered at pH 7.45 with either 100 mM Tris or 50 mM Tes, and the ionic strength was adjusted to 100 mM with NaCl. The concentrations of stock metal solutions were verified by ITC titrations against standardized EDTA solutions. A typical experiment consisted of a ∼0.25 mM or ∼1 mM metal solution titrated into an identically buffered ∼7 µM dimeric protein solution. Two or, more typically, three good data sets were collected for each type of titration, and the best-fit values were averaged and reported. The ITC data are presented as the baseline-adjusted raw data in the top panel and the peakintegrated, concentration-normalized heat of reaction versus the molar ratio of metal ion to protein dimer (or tetramer, where noted) in the bottom panel. The ITC data were fitted with the Origin software package provided by MicroCal (26), which uses a nonlinear leastsquares algorithm (minimization of χ2) and the concentrations of the titrant and the sample to fit the heat flow per injection to equations corresponding to an equilibrium binding model, providing best-fit values for the stoichiometry (nITC), change in enthalpy (∆HITC), and binding constant (KITC). Three different binding models, which have different assumptions and number of fit parameters, were used in this fitting, the one site model (3 parameters), the two independent sites model (6 parameters), and the multiple sequential sites model (2 parameters per site). The latter model imposes a sequential filling of individual binding sites and fixes the stoichiometry of each site at 1. Comparison of the goodness of fit with different models was based on the calculated χ2 value. The best-fit ∆HITC values for experiments in two or more buffers were then used to quantify the number of protons that are displaced upon metal binding. This is described in detail elsewhere (27, 28), and is based on the observation that the major difference between ∆HITC measured in two different buffers is due to the buffer protonation enthalpy (∆HbuffH+) times the number of protons (nH+) that are displaced; thus, a plot of ∆HITC versus ∆HbuffH+ can be used to determine nH+ at the experimental pH. Analysis of the experimental KITC (∆GITC) and ∆HITC values for the overall titration equilibrium in a thermodynamic cycle, consisting of ∆G° and ∆H° values for individual relevant equilibria, was then used to determine the buffer- and pH-independent log K (∆G°) and ∆H° values for metals binding to the proteins, as described elsewhere

10508 Biochemistry, Vol. 46, No. 37, 2007

Grossoehme et al. In addition, metal ions often displace protons when they bind to protein residues, and this leads to buffer protonation, which is another contribution to both KITC and ∆HITC. To account for these coupled protonation equilibria, the number of protons that are transferred at the experimental pH needs to be determined. As described in detail elsewhere (27, 28), the number of displaced protons can be quantified through an analysis of the experimental enthalpies (∆HITC) for two or more ITC titrations at the same pH with different buffers, and thus different buffer protonation enthalpies. Such an analysis of these ITC data in Tris and Tes buffers indicates that 0.90 proton is displaced when each Ni2+ binds to H144*UreE at pH 7.45. To extract the thermodynamic parameters (∆G°, ∆H°, ∆S°) for metal ion binding to proteins from the experimental ITC parameters, thermodynamic cycles were constructed for an analysis of the overall reaction that occurs upon addition of titrant. An example is shown in eqs 1-6 for Ni2+ binding to a H144*UreE subunit in Tris buffer, which is an appropriate cycle for Ni2+ binding to isolated sites on the subunits of the UreE dimer.

FIGURE 1: ITC data for the 25 °C titration of 0.25 mM Ni2+ into 6 µM H144*UreE dimer in 100 mM Tris, pH 7.45, and I ) 0.10 M. The upper panel contains the baseline-corrected raw data, and the lower panel indicates the peak-integrated, concentrationnormalized heats of reaction versus the molar ratio of Ni2+ per protein dimer. The solid line in the lower panel represents the best fit of the data using the one site model with nITC ) 1.92 ( 0.01, KITC ) 3.2 ( 0.2 × 106, ∆HITC ) -14.1 ( 0.1 kcal/mol.

(26, 27). Finally, the thermodynamic relationship ∆G° ) ∆H° - T∆S° was used to find the entropic contribution to binding. RESULTS AND ANALYSIS Ni2+ Binding to H144*UreE. ITC was used to quantify Ni2+ binding to H144*UreE from K. aerogenes, and Figure 1 shows a representative thermogram of Ni2+ titrated into the protein at pH 7.45 in 100 mM Tris buffer. A good fit to these data can be achieved using the one site model in the Origin software provided with the MicroCal titration calorimeter, and the average best-fit experimental values (nITC, KITC, ∆HITC) in Tris and Tes buffers are summarized in Table 1. In agreement with previous results (15, 16), the data clearly demonstrate that two (1.9 ( 0.1) Ni2+ ions bind to the protein dimer, but their affinities and enthalpies are indistinguishable by ITC, suggesting similar binding sites. Reverse titrations of H144*UreE into Ni2+ solutions (Figure S1 in Supporting Information) have an inflection at nITC ) 0.5 and are well fitted with similar values. Since ITC measures the total heat flow upon injection of titrant and titrations with metal ions typically have coupled equilibria involving the buffer and protonation, all of the equilibria that contribute to the experimental KITC and ∆HITC values need to be considered to determine ∆G° (Kd) and ∆H° for the equilibrium of interest (25). The two buffers were chosen not only for their buffering capacity at the desired pH but also for their formation of well-defined metal-buffer complexes (Table S1 in Supporting Information) that prevent metal hydrolysis reactions and allow the metal-buffer interactions to be accurately subtracted from KITC and ∆HITC.

xNiTris2+ + yNi(Tris)22+ + (1 - x - y)Ni2+ + (1 - r)H144*UreE + rH144*UreE-H+ + (r - x - 2y)Tris h H144*UreE-Ni2+ + rTrisH+ (1) x{NiTris2+ h Ni2+ + Tris}

(2)

y{Ni(Tris)22+ h Ni2+ + 2Tris}

(3)

r{H144*UreE-H+ h H144*UreE + H+}

(4)

Ni2+ + H144*UreE h H144*UreE-Ni2+

(5)

+

+

r{H + Tris h TrisH }

(6)

The overall reaction for the formation of H144*UreE-Ni2+ is given by eq 1, where x ) 0.35, y ) 0.60, and r ) 0.90 are coefficients that are fixed by the experimental conditions (pH 7.45, 100 mM Tris) and indicate the different Ni2+ species (60% Ni(Tris)22+, 35% NiTris2+, 5% Ni2+) and the H144*UreE protonation states relevant to the Ni2+ binding reaction (90% H144*UreE-H+, 10% H144*UreE) that are present in solution. The cycle is then constructed from the relevant weighted individual equilibria (eqs 2-6) that sum to the overall equilibrium (eq 1) and allow the buffer- and pH-independent values of ∆G° (K) and ∆H° for Ni2+ binding to the protein (eq 5) to be determined (e.g., ∆HITC ) x∆H°eq2 + y∆H°eq3 + r∆H°eq4 + ∆H°eq5 + r∆H°eq6). Pertinent ∆G° (K1, β2) and ∆H° values for metal-buffer interaction (e.g., eqs 2, 3), including several determined as part of this study, are found in Table S1 (Supporting Information). Other necessary values are those associated with the 0.90 proton that Ni2+ displaces upon binding to the protein (eq 4) and the subsequent buffer protonation (eq 6), for which pKa and ∆H° values are known. Since NMR and EXAFS data indicate that each bound Ni2+ has two His ligands, the displaced 0.90 proton is likely to be associated with these residues and eq 4 was modeled2 with pKa ) 8.40 and the His deprotonation enthalpy (7.10 kcal/mol (29)). Analogous thermodynamic cycles that include ∆G° and ∆H° values for the relevant metal-buffer species and protein protonation states under the experimental conditions were

Thermodynamics of Metal Binding to UreE

Biochemistry, Vol. 46, No. 37, 2007 10509

Table 1: Summary of Average Best-Fit Valuesa from 25 °C ITC Data for UreE Dimer Titrations, the Corresponding pH- and Buffer-Independent Values, and Idealized Total Metal-Dimer Ratios. Ni2+

protein

buffer

nITCb

KITC

∆HITC (kcal/mol)

log K

∆H° (kcal/mol)

total M2+/dimer

H144*UreEc

Tris Tes Tris

1.9 (0.1) 2.2 (0.1) 2 3.3 (0.2) 2 3.1 (0.4) 1 1 1 1 2.1 (0.1) 2.2 (0.1) 2.2 (0.1)

2.8 ( 0.4 × 106 2.6 ( 0.3 × 106 1.0 ( 0.5 × 106 1.0 ( 0.1 × 105 1.7 ( 0.7 × 106 7 ( 1 × 104 1.3 ( 0.4 × 105 3 ( 1 × 103 1.2 ( 0.4 × 105 1.6 ( 0.7 × 104 4.5 ( 0.2 × 105 1.2 ( 0.1 × 106 1.3 ( 0.1 × 105 9.8 ( 0.7 × 103

-14.1 (0.2) -14.9 (0.1) -13.3 (0.8) -5.8 (0.5) -13.3 (0.4) -5 (1) -3.3 (0.4) -2 (3) -3.2 (0.5) -1 (1) -9.2 (0.1) -9.6 (0.1) -11.7 (0.3)

8.8 (0.2) 8.8 (0.1)

-14 (2) -14 (4)

2 2 2 5-6 2 5-6 1 2 1 2 2 2 2

KaUreEd

Tes BpUreEe

Tris Tes

Cu2+

H144*UreEc

Zn2+

H144*UreEc BpUreEf

Tris Tes Tris Tris

6.98 (0.08)

-5 (2)

6.89 (0.09) 7.55 (0.08) 6.2 (0.2) 7.5 (0.2) 6.7 (0.2) 10.8 (0.5) 11.3 (0.2) 7.06 (0.08)

-5 (3) -3 (1) 0 (4) -2 (1) 2 (2) -10 (4) -10 (4) -11 (2)

a Values are from fits with the one site model, unless indicated otherwise; the associated error is shown in parentheses. b Integer value is a fixed fit parameter. c Values for H144*UreE were obtained with low protein dimer concentrations (